基于数据挖掘的中文理工科纸本图书采访经费分配模型

Book Acquisition Budget Allocation Model of Science and Engineering Disciplines Based on Data Mining

以复旦大学图书馆2005-2010年的流通数据、复旦大学院系师生人数、2005-2010年新闻出版业出版情况为数据源,通过数据筛选、清洗、处理等方法,建模并进行多元回归分析,从海量实际数据中挖掘出影响采访经费分配的显著变量,确定变量之间的数学关系式,并进行统计校验,由此得出采访经费比例预测模型。

This paper describes the book acquisition budget allocation model based on data mining, which analyzes millions of circulation data, the number of faculty and students of various disciplines, press and publication data in China from 2005 to 2010. The methods of the data sieving ad data cleaning are used to process above mass data. This paper mines significant variables of book acquisition budget allocation with multiple regression analysis. Finally, an optimized allocation prediction model is established and tested.